Primary structure of elongation factor Tu from

Proc. Natl. Acad. Sci. USA Vol. 77, No. 3, pp. 1326-1330, March 1980 Biochemistry

Primary structure of elongation factor Tu from Escherichia coli (sequence heterogeneity/active sites/protein homology/posttranslational modification/prediction of secondary structure)

K. ARAI*, B. F. C. CLARKt, L. DUFFYt, M. D. JONESt, Y. KAZIRO*, R. A. LAURSEN*, J. L'ITALIENt, D. L. MILLER"1, S. NAGARKATTIt, S. NAKAMURA*, K. M. NIELSENt, T. E. PETERSENt, K. TAKAHASHII1, AND M. WADES *Institute of Medical Science,.University of Tokyo, Minatoku, Tokyo, Japan; tDepartments of Chemistry and Molecular Biology, Aarhus University, Aarhus, Denmark; tChemistry Department, Boston University, Boston, Massachusetts 02215; §Roche Institute of Molecular Biology, Nutley, New Jersey 07110; and I'Primate Research Institute, Kyoto University, Inuyama, Japan Communicated by B. L. Horecker, December 5, 1979

The amino acid sequence of elongation factor ABSTRACT Tu (EF-Tu) from Escherichia coli has been determined. EF-Tu is a single-chain polypeptide composed of 393 amino acids (M, 43,225 for the species bearing COOH-terminal serine). The NHrterminal serine is acetylated, and lysine-56 is partially methylated. The sites of facile tryptic cleavage are at arginines 44 and 58 and at lysine-263. The cysteinyl residues associated with aminoacyl-tRNA and guanosine nucleotide binding activities are residues 81 and 137, respectively. The COOH-terminal amino acid is heterogeneous in that analyses of the COOH-terminal peptides isolated from different EF-Tu preparations gave position 393 as glycine and serine in ratios (Gly/ Ser) ranging from about 0.7 to 3.

Elongation Factor Tu (EF-Tu) promotes the binding of aminoacyl-tRNA (AA-tRNA) to ribosomes during protein biosynthesis in Escherichia coli (1-3). The protein consists of a single polypeptide chain of Mr 43,225 (for the species bearing COOH-terminal serine). Proteins of similar size and function are' essential constituents of other prokaryotes and also of eukaryotes (4, 5). The protein is interesting not only because of this important biological function but also because of the number and diversity of the substances with which it interacts. In addition to AA-tRNA, the protein binds GDP, GTP, EF-Ts, the antibiotics kirromycin (6), Aurodox (X5108) (7), pulvomycin (8), puromycin (P. Grant and B. Cooperman, personal communication), and certain as-yet-undefined components of the ribosome. In addition to promoting peptide chain elongation, EF-Tu is a subunit of bacteriophage QB RNA polymerase

(9).

During peptide chain elongation EF-Tu binds GTP before combining with AA-tRNA to form the ternary complex AAtRNA-EF-Tu-GTP. The ternary complex will bind to ribosomes containing the appropriate codon in the A (AA-tRNA binding) site; then GTP is hydrolyzed, EF-Tu-GDP is released, and AA-tRNA is oriented on the ribosome so that peptide bond formation can occur. The role of GTP may be described as that of an allosteric effector, which alters the'protein's tertiary structure to expose AA-tRNA binding sites (3, 10). Conversely, the binding of GDP may mask these sites, because the affinity constant of the EF-Tu-GDP for AA-tRNA differs from that of EF-Tu-GTP by a factor of about 105 (11). This cyclic opening and closing of the AA-tRNA binding sites dependent upon the state of phosphorylation may be regarded as a simple energy transduction system and is reminiscent of contractile proteins. As the foundation for detailed studies of the structure and function of EF-Tu and its complexes, we have determined its

primary structure. This report describes the sequence determined independently in three laboratories. Because the experimental approaches employed by our laboratories differed substantially, full details will be reported elsewhere. MATERIALS AND METHODS EF-Tu was purified as the GDP complex from E. coli B, E. coli Q13, or E. coli MRE 600, according to published procedures (12, 13). So far we have detected no differences in the sequences of EF-Tu isolated from these strains. Peptides were generated by cleavage with cyanogen bromide or limited trypsinolysis as described (14-17); large peptides were further fragmented with trypsin, Staphylococcus aureus protease, or chymotrypsin. Amino acid sequences were determined either by the manual dansyl-Edman procedure (18-23) or by automated solid-phase Edman degradation (24, 25). Phenylthiohydantoins were

identified by thin-layer chromatography and high-performance liquid chromatography, or by back hydrolysis to amino acids. RESULTS AND DISCUSSION The amino acid sequence of EF-Tu is shown in Fig. 1. The amino acid composition calculated from the sequence and listed in Table 1 is in good agreement with the compositions previously published (26, 27), if corrections are made for the differing molecular weights. The protein contains two modifications: Lys-56 is methylated (28), and Ser-1 is acetylated. ** The amino acid composition of EF-Tu differs in few respects from the average for' 108 protein families compiled by Dayhoff and Hunt (29). EF-Tu contains about 20% more charged residues and 30% more aliphatic hydrophobic residues than the average. The uncharged hydrophilic residues are correspondingly reduced, there being 51% fewer Asn + Gln and 61% fewer Ser than expected. The ratio of Thr to Ser is 2.7, whereas the average ratio is 0.85. These residues are generally interchangeable, and the Ser deficit is partially compensated for by a Thr excess. Acidic residues outnumber basic ones. Assuming half the His residues are protonated, the protein has a net negative charge at neutral pH of about -10, befitting a protein whose isoelectric point is 5.5. The distributiQn of residues along the peptide chain (Fig. 2) shows some patterns possibly related to the protein's structure and function. The first 119 residues form a region that is highly hydrophilic and basic; 8 of the 11 His residues lie here. Histidine has been implicated in AA-tRNA and GDP binding (30); furthermore, this region contains not only the Cys residues whose Abbreviations: EF, elongation factor: AA-tRNA, aminoacyl-tRNA. 1 To whom reprint requests should be addressed. ** M. D. Jones, T. E. Petersen, and K. M. Nielsen, unpublished data.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact. 1326

Biochemistry:

Proc. Natl. Acad. Sci. USA 77 (1980)

Arai et al.

1327

30 20 10 1 Acetyl -Ser-Lys-Gl u-Lys-Phe-Gl u-Arg-Thr-Lys- Pro-Hi s-Val -Asn-Vaal -Gly-Thr- IIe-Gly-Hi s-Val -Asp-Hi s-Gly-Lys-Thr-Thr-Leu-Thr-Al a-Al a60 50 (Me) 40 Il e-Thr-Thr-Val -Leu-Al a-Lys-Thr-Tyr-Gly-Gly-Al a-Al a-Arg-Al a-Phe-Asp-Gl n-Il e-Asp-Asn-Ala-Pro-Gl u-Gl u-Lys-Al a-Arg-Gly- Ile90 80 70 Thr- Il e-Asn-Thr-Ser-Hi s-Val -Gl u-Tyr-Asp-Thr-Pro-Thr-Arg-His-Tyr-Al a-Hi s-Val -Asp-Cys-Pro-Gly-His-Al a-Asp-Tyr-Val -Lys-Asn-

100

110

120

Met-Il e-Thr-Gly-Al a-Al a-Gl n-Met-Asp-Gly-Al a-Il e-Leu-Val -Val -Al a-Al a-Thr-Asp-Gly-Pro-Met-Pro-Gl n-Thr-Arg-Gl u-His-Ile-Leu130

140

150

Leu-Gly-Arg-Gln-Val-Gly-Val-Pro-Tyr-Ile-Ile-Val-Phe-Leu-Asn-Lys-Cys-Asp-Met-Val -Asp-Asp-Glu-Glu-Leu-Leu-Glu-Leu-Val-Glu160

170

180

Met-Glu-Val-Arg-Glu-Leu-Leu-Ser-Gln-Tyr-Asp-Phe-Pro-Gly-Asp-Asp-Thr-Pro-Ile-Val -Arg-Gly-Ser-Ala-Leu-Lys-Ala-Leu-Glu-Gly210 200 190 Asp-Al a-Gl u-Trp-Gl u-Al a-Lys- Il e-Leu-Gl u-Leu-Al a-Gly-Phe-Leu-Asp-Ser-Tyr- Ile-Pro-Gl u-Pro-Gl u-Arg-Ala-Ile-Asp-Lys-Pro-Phe240 230 220 Leu-Leu-Pro- Ii e-Gl u-Asp-Val-Phe-Ser- Il e-Ser-Gly-Arg-Gly-Thr-Val-Val-Thr-Gly-Arg-Val -Gl u-Arg-Gly- Ile-Ile-Lys-Val-Gly-Gl u-

270 260 250 G1 u-Val-Gl u-Il e-Val -Gly- Il e-Lys-Gl u-Thr-Gl n-Lys-Ser-Thr-Cys-Thr-Gly-Val-Glu-Met-Phe-Arg-Lys-Leu-Leu-Asp-Gl u-Gly-Arg-Al a280

290

300

Gly-Gl u-Asn-Val -GI y-Val -Leu-Leu-Arg-Gly- I 1 e-Lys-Arg-Gl u-Gl u- Il e-G1 u-Arg-Gly-Gl n-Val -Leu-Al a-Lys-Pro-Gly-Thr- I i e-Lys-Pro330 320 310 Hi s-Thr-Lys-Phe-Gl u-Ser-Gl u-Val -Tyr- II e-Leu-Ser-Lys-Asp-Glu-Gly-Gly-Arg-His-Thr-Pro-Phe-Phe-Lys-Gly-Tyr-Arg-Pro-Gl n-Phe340 360 350 Tyr-Phe-Arg-Thr-Thr-Asp-Val-Thr-Gly-Thr- II e-Gl u-Leu-Pro-Gl u-Gly-Val --Glu-Met-Val -Met-Pro-Gly-Asp-Asn- Il e-Lys-Met-Val -Val -

370 390 380 Thr-Leu- I1e-Hi s-Pro- I1 e-Al a-Met-Asp-Asp-Gly-Leu-Arg-Phe-Al a- Ile-Arg-Gl u-Gly-Gly-Arg-Thr-Val -Gly-Al a-Gly-Val -Val -Al a-LysVal -Leu-Ser-OH- (Gly-OH)

FIG. 1. Amino acid sequence of EF-Tu from E. coli. The COOH-terminal residue is heterogeneous, both Ser and Gly being found.

modification inhibits the interaction with AA-tRNA, guanosine nucleotides, and EF-Ts, but also the sites of facile tryptic cleavage that affect AA-tRNA binding. The central region, residues 119-250, contains the highest densities of both aliphatic hydrophobic and acidic residues in the protein: 42% of the hydrophobic residues and 46% of the Table 1. Amino acid composition of EF-Tu

Residue

Residues per molecule From analysis From sequence

23 21-25 Lys 23 24 Arg 11 His 10-11 24 34-36 Asp Asn 7 Glu 37 49-50 Gln 8 11 Ser 11 Thr 30 30-31 3 3 Cys Pro 20 19-20 40 Gly 40-43 Ala 27 28 Val 37 33-34.5 Leu 28 29-31 Ile 29 25-28 Met 10 7.4-9.1 Phe 14 14 10 Tyr 9 1 Trp 2-3 The sequence composition refers to the species bearing COOHterminal Ser. The analyses for Asp or Glu include Asn or Gln. Analytical data from refs. 15 and 16 were corrected to a Mr of 43,225.

acidic residues are located here. Together these six amino acids account for 55% of the residues in this region. The ratio of acidic to basic residues is 2.0. The COOH-terminal region, residues 251-393, like the NH2-terminal region, is rich in basic residues but in contrast has a compensating number of acidic residues. The fraction of hydrophobic residues in this region is 10% lower than that of total protein. Sequence Heterogeneity. Because there are two genes for EF-Tu in the E. coli chromosome (31), the protein preparation whose sequence was determined may consist of two gene products differing in primary structure. It is not likely, however, that these gene products differ extensively in sequence. Previous studies showed that the two gene products differed only slightly in their activities and in their tryptic peptide maps (32, 33), and the DNA from the genes hybridized completely under stringent conditions (34). No sequence heterogeneity has yet been clearly identified except for the COOH-terminal residue 393, where variable amounts of glycine and serine have been found in different digests. The COOH-terminal CNBr fragment from E. coli B was found to contain about 0.4 Ser residues, and from a tryptic digest of this fragment, two peptides, Val-Leu-Gly and ValLeu-Ser, were isolated in a molar ratio of 4:1. In other preparations the ratio of Gly to Ser in mixtures of COOH-terminal tryptic peptides has varied from 0.7 to 2. Hydrazinolysis of intact EF-Tu preparations released both Gly and Ser in ratios of 2.3 and 2.9 (Gly/Ser). On the other hand, carboxypeptidase Y digestion of EF-Tu isolated from the MRE 600 strain released only Ser in significant quantities (17). Whether this heterogeneity results from the two gene products will only be known when their DNA sequences are determined.

Biochemistry:

1328

Arai et al.

Proc. Natl. Acad. Sci. USA 77 (1980) Residue

Lys

50

25

1

I

I

I

75

I

100

125

150 1

I

175 . I

225 *

200 *

I

II I I I I I I 11 I II I Asp I 11 I 11 I I 11 11 I Glu 11 I 11111 1 111 I Asn I GIn I I II Ser I Thr l 1111 I II li1 Arg His

I

I

I

I I I II

250 I

I

11

11

275 I I I I

I

III I II 11

I

325

300

I III III

I

I

375

I I

11

I

I 11 I

350

I

I

III

I

I I I I

I I II~~~~~~~~~~~ 11 I III 11

II I II III I Cys I I Pro I I I II I 11 I 1111 II 11 II I Gly I II I 11111 I III I I II I I 11 I liii II II I I 11111 Ala H il 1 1111 11 I I lI I 11 Val I III I11 I II I III I 1 11111 I 11 I I 11 I III Leu I I 11 III I I 11 I I I II I I I III lle I I I I 11 I I I 1111 I I I11 II I I l Iii Met II I I I I l II I I Phe I I II I I 11 I I Tyr I II I 11 11 I Trp

Structure

L

i

FIG. 2. Amino acid distribution in EF-Tu. Vertical bars indicate sequence position of each amino acid. Predicted regions of secondary structure are shown by helices (a helix) and zig-zags (f3 structure).

Active Sites and Trypsin Cleavage Sites. Residues that had previously been identified by chemical modification can now be located in the complete sequence. The Cys that has been implicated in the AA-tRNA binding function (35,36) is located at position 81, and the Cys whose modification by N-ethylmaleimide inhibits the binding of GDP, GTP, and EF-Ts is at position 137. Cys-255 has not been modified in the native protein and is probably buried in the three-dimensional structure.

Trypsin rapidly cleaves EF-Tu at Arg-44 and Arg-58 to give large fragment (A) and two small fragments (E and F) with concomitant loss of AA-tRNA binding activity (16, 37). Intermediates (A' and D) formed by cleavage at either residue have also been observed (38). In a longer incubation trypsin cleaves fragment A at Lys-263, forming fragments B and C (16, 17). It has been observed that brief trypsin treatment enhances the formation of well-ordered crystals suitable for x-ray diffraction (39-41). These crystals contain fragments A and D, or A and E, the latter crystals presumably lacking fragment F. The structures and sizes of these fragments are listed in Table 2. a a

Table 2. Sizes of tryptic fragments of EF-Tu

Fragment designation A' A B C D E F

Residues 45-393

Mr

37,849 35,971 19,324 16,647 7,254 5,376 45-58 1,878 Fragments are produced by brief treatment of EF-Tu with trypsin. The nomenclature for fragments A-D is that of Nakamura et al. (17). 59-393 59-263 264-393 1-58 1-44

At present, no direct chemical information about the guanosine nucleotide binding site is available. An analysis of the

electron density maps obtained by x-ray crystallography reveals GDP binding site associated with a region of the molecule containing a high proportion of secondary structure (41). Currently, a chain tracing based on the sequence in this report is being pursued, but it is not yet clear which amino acid residues are involved in GDP binding (T. LaCour, J. Nyborg, J. Rubin, and B. F. C. Clark, unpublished). Posttranslational Modifications. One of the first observations on the primary structure of EF-Tu was that the NH2terminal residue was blocked. This terminal residue has been identified as N-acetylserine. ** The sequence of the tufA DNA coding for this region of the protein reveals that the serine codon (UCU) is preceded by GUG (T. Yokota, H. Sugisaki, M. Takanami, and Y. Kaziro, unpublished data); thus translation initiates with formylmethionine, which then is cleaved and the resulting NH2-terminal serine is acetylated. Ames and Nikaido (42) found that EF-Tu is one of a very few prominent proteins in E. coli cytosol methylated by methionine. Both monomethyl- and dimethyllysine were formed. It has been found that essentially all of the methylation occurs at Lys-56 (28). In the protein preparations used for sequence determination, 43% of Lys-56 is monomethylated and 7% is dimethylated. This region of the protein is probably exposed because here both trypsin cleavage and methylation occur preferentially. Structural Relationships Between EF-Tu and Other Proteins. In an attempt to understand the function and evolution of EF-Tu, comparisons have been made to other proteins, notably actin (43). Actin is also selectively cleaved by trypsin at an Arg-Gly bond (44). Nevertheless, aside from the common Arg-Gly-Ile sequence at the trypsin cleavage site (58-60 in EF-Tu; 62-64 in actin) and a Lys-Cys-Asp sequence (136-138 in EF-Tu; 283-285 in actin), no homologies have been detected. a

Biochemistry:

Proc. Natl. Acad. Sci. USA 77 (1980)

Arai et al.

Table 3. Predicted regions of secondary structure a helix f sheet Residues P3 Residues Pa Pay PI, 14-20 0.88 0.81 1.22 1.18 2-9 1.17 0.95 31-35 1.00 1.39 42-50 60-64 0.90 1.29 0.61 1.26 52-57 95-107* 138-157 174-192 203-208 240-245 258-267 303-315 372-378

1.20 1.26 1.26 1.19 1.29 1.22 1.12 1.19

1.08 0.92 0.87 0.84 1.02 0.97 0.93 1.02

102-108* 124-134 274-281 329-341 358-363 387-392*

1.15 0.96 0.97 0.99 1.11 1.16

1.31 1.35 1.26 1.20 1.40 1.33

* Region of similar a and : probabilities.

A much stronger homology has been discovered between EF-Tu and EF-G (45), proteins which in addition to having related functional roles are located adjacent to one another in the E. coli chromosome (46). Like EF-Tu, EF-G contains a site of preferential trypsin cleavage (47). If the major fragments are aligned at their cleavage sites, seven out of the first eight amino acids are identical; for the first 60 amino acids, the degree of identity is 33%. A thorough assessment of the homology must await the determination of the remaining 80% of the EF-G sequence. The functional resemblances among initiation factor 2, EF-Tu, EF-G, and release factor suggest that they may have descended from a common ancestor or have acquired some sequence similarity through convergent evolution. Other likely relatives of EF-Tu are the eukaryotic, mitochondrial, and chloroplast elongation factors. Confirmation by DNA Nucleotide Sequence Determination. Much of the protein sequence has now been confirmed by determining the sequence of the DNA of the Sma I restriction fragment containing the tufA gene, which lies adjacent to the gene for EF-G in the E. coli chromosome (46). The DNA sequence, to be reported elsewhere (T. Yokota, H. Sugisaki, M. Takanami, and Y. Kaziro, unpublished data), confirms the protein sequence between residues 1-18 and 28-393.Jt The agreement between the protein sequence and the tufA DNA sequence reinforces our conclusion that the protein preparation has a homogenous sequence despite the possible presence of two gene products. Secondary Structure Predictions. Extending and updating an earlier report (15), we estimated the regions of a helix and f3 sheet according to the method of Chou and Fasman (Table 3) (48). The average conformation parameters (Pa and Pa) were calculated by using the revised single-residue parameters based upon 29 proteins (49). EF-Tu was calculated to contain 30% a helix, in good agreement with the value of 24-32% obtained by circular dichroism measurements (2, 50). The /3 sheet content was calculated to be 16%. The a and 3 structures are not evenly distributed throughout the protein. The central region of the molecule, trypsin fragment B, is composed of 34% a helix and 8% /3 sheet, whereas the COOH-terminal third of the molecule is 15% a helix and 25% 3 sheet. The relationship of the secondary structure to the protein's function is not yet apparent. The region 44-57, which is thought to be involved in AA-tRNA binding, contains two helical segments; however, because the binding of GDP or GTP does not alter the circular dichroism of the protein, neither nucleotide is likely to induce a major change in helix content. ff Since submission of this manuscript we have found that the codon

for the COOH-terminal amino acid is GGC (glycine).

1329

In an overall view of the protein based upon its primary structure and the predictions of its secondary structure, EF-Tu is composed of a basic, hydrophilic, NH2-terminal region composing one-third of the molecule and containing all of the residues thus far implicated in its function. The central region of the molecule is rich in acidic and hydrophobic residues arranged in a helices. This region might interact with basic ribosomal proteins. The COOH-terminal region has no known function but may well contain part of the AA-tRNA and GDP binding sites. With the aid of the primary structure, many of the gaps in our understanding of the function of EF-Tu can now be filled by using x-ray crystallography and chemical modification. 1. Lucas-Lenard, J. & Lipmann, F. (1971) Annu. Rev. Biochem. 40,409-448. 2. Miller, D. L. & Weissbach, H. (1977) in Molecular Mechanisms of Protein Biosynthesis, eds. Weissbach, H. & Pestka, S. (Academic, New York), pp. 323-373. 3. Kaziro, Y. (1978) Biochim. Biophys. Acta 505,95-127. 4. Iwasaki, K., Motoyoshi, K., Nagata, S. & Kaziro, Y. (1976) J. Biol. Chem. 251, 1843-1845. 5. Slobin, L. T. & Moller, W. (1976) Eur. J. Biochem. 69, 351366. 6. Wolf, H., Chinali, G. & Parmeggiani, A. (1974) Proc. Natl. Acad. Sci. USA 71, 4910-4914. 7. Wilson, G. E. & Cohn, M. (1977) J. Biol. Chem. 252, 20042009. 8. Wolf, H., Assmann, D. & Fischer, E. (1978) Proc. Natl. Acad. Sci. USA 75, 5324-5328. 9. Blumenthal, T. & Carmichael, G. G. (1979) Annu. Rev. Biochem. 48,525-548. 10. Crane, L. J. & Miller, D. L. (1974) Biochemistry 13,933-939. 11. Shulman, R. G., Hilbers, C. W. & Miller, D. L. (1974) J. Mol. Biol.

90,601-607. 12. Miller, D. L. & Weissbach, H. (1974) Methods Enzymol. 30, 219-232. 13. Arai, K., Kawakita, M. & Kaziro, Y. (1972) J. Biol. Chem. 247, 7029-7037. 14. Wade, M., Laursen, R. A. & Miller, D. L. (1975) FEBS Lett. 53, 37-39. 15. Laursen, R. A., Nagarkatti, S. & Miller, D. L. (1977) FEBS Lett.

80,103-106. 16. Nakamura, S., Arai, K., Takahashi, K. & Kaziro, Y. (1975) Biochem. Biophys. Res. Commun. 66, 1069-1077. 17. Nakamura, S., Arai, K., Takahashi, K. & Kaziro, Y. (1977) Biochem. Biophys. Res. Commun. 77, 1418-1424. 18. Welner, A. M., Platt, T. & Weber, K. (1972) J. Biol. Chem. 247, 3242-3251. 19. Van Eerd, J.-P. & Takahashi, K. (1976) Biochemistry 15, 1171-1180. 20. Bennet, J. C. (1967) Methods Enzymol. 11, 330-339. 21. Gray, W. R. (1967) Methods Enzymol. 11, 469-475. 22. Vanderkerckhove, J. & Van Montagu, M. (1974) Eur. J. Biochem.

44,279-288. 23. Peterson, J. D., Nierhlich, S., Oyer, P. E. & Steiner, D. F. (1972)

J. Biol. Chem. 247,4866-4871. 24. Laursen, R. A. (1971) Eur. J. Biochem. 20,89-102. 25. Laursen, R. A. (1977) Methods Enzymol. 47,277-288. 26. Miller, D. L. & Weissbach, H. (1970) Arch. Biochem. Biophys. 141,26-37. 27. Arai, K., Kawakita, M., Kaziro, Y., Kondo, T. & Ui, N. (1973) J. Biochem. (Tokyo) 73, 1095-1105. 28. L'Italien, J. J. & Laursen, R. A. (1979) FEBS Lett. 107, 359362. 29. Dayhoff, M. 0. & Hunt, L. T. (1972) in Atlas of Protein Sequence and Structure, ed. Dayhoff, M. O. (National Biomed. Res. Fdn., Washington, DC), Vol. 5, p. D-355. 30. Nakano, A., Miyazawa, T., Nakamura, S. & Kaziro, Y. (1979)

Arch. Biochem. Biophys. 196,233-238.

Jaskunas, S. R., Lindahl, L., Nomura, M. & Burgess, R. R. (1975) Nature (London) 257,458-462. 32. Furano, A. (1977) J. Biol. Chem. 252, 2154-2157. 31.

1330

Biochemistry: Arai et al.

33. Miller, D. L., Nagarkatti, S., Laursen, R. A., Parker, J. & Friesen, J. D. (1978) Mol. Gen. Genet. 159,57-62. 34. Furano, A. (1978) Proc. Natl. Acad. Sci. USA 75,3104-3108. 35. Miller, D. L., Hachmann, J. & Weissbach, H. (1971) Arch. Biochem. Biophys. 144, 115-121. 36. Arai, K., Kawakita, M., Nakamura, S., Ishikawa, I. & Kaziro, Y. (1974) J. Biochem. (Tokyo) 76,523-534. 37. Arai, K., Nakamura, S., Arai, T., Kawakita, M. and Kaziro, Y. (1976) J. Biochem. (Tokyo) 79, 69-83. 38. Blumenthal, T., Douglass, J. & Smith, D. (1977) Proc. Natl. Acad. Sci. USA 74,3264-3267. 39. Gast, W. H., Kabsch, W., Wittinghofer, A. & Leberman, R. (1977) FEBS Lett. 74, 88-90. 40. Jurnak, F., Rich, A. & Miller, D. (1977) J. Mol. Biol. 115, 103110. 41. Morikawa, K., La Cour, T. F. M., Nyborg, J., Rassmussen, K. M., Miller, D. L. & Clark, B. F. C. (1978) J. Mol. Biol. 125, 325338.

Proc. Nat!. Acad. Sci. USA 77 (1980) 42. Ames, G. F.-L. & Nikaido, K. (1979) J. Biol. Chem. 254, 9947-9950. 43. Rosenbusch, J. P., Jacobson, G. R. & Jaton, J.-C. (1976) J. Supramol. Struct. 5,391-396. 44. Jacobson, G. R. & Rosenbusch, J. P. (1976) Proc. Natl. Acad. Sci. USA 73,2742-2746. 45. Laursen, R. A. & Duffy, L. (1978) FEBS Lett. 92, 200202. 46. Nomura, M., Morgan, E. A. & Jaskunas, S. R. (1977) Annu. Rev. Genet. 11,297-347. 47. Alakhov, Y. B., Motuz, L. P., Strengrevics, 0. A. & Ovchinnikov, Y. A. (1978) FEBS Lett. 85, 287-290. 48. Chou, P. Y. & Fasman, G. D. (1974) Biochemistry 13, 222245. 49. Chou, P. Y. & Fasman, G. D. (1978) Adv. Enzymol. Relat. Areas Mol. Biol. 47,45-148. 50. Ohta, S., Nakanishi, M., Tsuboi, M., Arai, K. & Kaziro, Y. (1977) Eur. J. Biochem. 78,599-608.